RE: Spec review: USB Client

2007-06-28 Thread Lynch, Rusty
Frank, the Ubuntu Mobile project isn't going to be writing any Windows drivers. 
 I would suggest that if other operating systems have issues with the proposed 
implementation for providing "USB Client", then this should be highlighted in 
outstanding issues or even include that in the rational for choosing one 
possible design over another.

Ok, about the current content in the spec... we seem to be badly missing some 
kind of a big picture type of discussion at the top of the page, making the 
couple of bullets in Use Cases and then the diagram in Design really confusing.

The Summary should give some introduction on how this device is expected to 
interact (at a high level, not jumping into topics like the USB Communications 
Device Class (CDC) standard), let the Use Cases flesh out a lot more of the 
interaction details, and in the design dig into the usb gadget stuff.  I think 
the debate that Frank is wanting get into can be productive once the potential 
use cases have been fleshed out.

Hmmm... just thinking out loud for a moment on use cases... 

<>
User connects a USB cable from the client USB port of the device to the host 
USB port of another computer:

The device wakes up and opens a dialog asking if the user would like to share a 
network connection, share personal files on the device, or ignore the 
connection.

<>
User selects the "share a network connection" option

X

<>
User selects the "share personal files" option

A full screen dialog opens on the device letting the user know that the device 
is in file share mode, and that the cancel button must be pressed in order to 
cancel sharing files and start using the device again.

On the other computer, a new mass storage device is announced on the USB bus.  
The expectation is that the other computer will auto mount the new device, and 
any specialty applications written for the device will respond to the new mount 
and take appropriate actions (in the same way most modern laptop's will respond 
to plugging in a camera that exposes a mass storage class interface over USB.)

If the other computer is another device running UME, then the second system 
will wake (if sleeping), automount the mass storage device provided by the 
first system, and open a file sharing application that will allow the user to 
select video, music, and pictures to copy between the two systems.

<> 
User selects the "Ignore" option

The device continues to operate as if the USB cable was not connected, other 
then the device begins to charge (if the device supports charging over USB) and 
software on the system response appropriately to the power management message 
that the system is charging.  (For example, the battery status applet would be 
updated for the new power state.)


... ok, hopefully people see where I am going with this.  I need to get some 
sleep to get ready for a fun filled day at OLS.

--rusty

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Li, Frank
Sent: Thursday, June 28, 2007 7:00 PM
To: Tollef Fog Heen; ubuntu-mobile@lists.ubuntu.com
Subject: RE: Spec review: USB Client

Toll:
I have some question about USB Client. 
1.  I remember WINDOWS XP doesn't provide CDC driver, windows xp sp2 just 
provide RNDIS driver for usb to eth. 
Do you prepare to develop windows CDC driver?
2.  Mass storage device has problem between two systems. Device just 
provide umounted storage device to host by Mass storage device. Mass storage 
device use SCSI command over USB. So it can't resolve share problem when device 
and host access the same file. When host copy a file to device, device will no 
see such file before remount the device because file system uses buffer. 
I suggest use SMB or NFS. 

Best regards
Frank Li

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tollef Fog Heen
Sent: 2007年6月28日 20:32
To: ubuntu-mobile@lists.ubuntu.com
Subject: Spec review: USB Client


I am currently in the process of going through the various
specifications submitted for review or proposal.  Since I suspect we
will end up with a couple of emails back and forth, I'll be splitting
those into separate emails.

This email is about the USB client one,
https://wiki.ubuntu.com/MobileAndEmbedded/USBClient

- The release note bit is not filled in, this needs to happen.

- Some boilerplate text is still present in other sections.  Please
  remove anything which isn't applicable to the spec, or fill in the
  section.

The specification lists a couple of different items that need to be
developed:

- Linux utility

- CDC Ethernet

- File-backed storage

- Peripherial controller driver

The spec should at least list how those bits interact and more
importantly, what other features in the OS they hook into and if those
need any changes.  Ideally, a spec should be implementable by a person
who only has the spec as his or her guidance as to how to implement a

RE: Spec review: USB Client

2007-06-28 Thread Li, Frank
Toll:
I have some question about USB Client. 
1.  I remember WINDOWS XP doesn't provide CDC driver, windows xp sp2 just 
provide RNDIS driver for usb to eth. 
Do you prepare to develop windows CDC driver?
2.  Mass storage device has problem between two systems. Device just 
provide umounted storage device to host by Mass storage device. Mass storage 
device use SCSI command over USB. So it can't resolve share problem when device 
and host access the same file. When host copy a file to device, device will no 
see such file before remount the device because file system uses buffer. 
I suggest use SMB or NFS. 

Best regards
Frank Li

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tollef Fog Heen
Sent: 2007年6月28日 20:32
To: ubuntu-mobile@lists.ubuntu.com
Subject: Spec review: USB Client


I am currently in the process of going through the various
specifications submitted for review or proposal.  Since I suspect we
will end up with a couple of emails back and forth, I'll be splitting
those into separate emails.

This email is about the USB client one,
https://wiki.ubuntu.com/MobileAndEmbedded/USBClient

- The release note bit is not filled in, this needs to happen.

- Some boilerplate text is still present in other sections.  Please
  remove anything which isn't applicable to the spec, or fill in the
  section.

The specification lists a couple of different items that need to be
developed:

- Linux utility

- CDC Ethernet

- File-backed storage

- Peripherial controller driver

The spec should at least list how those bits interact and more
importantly, what other features in the OS they hook into and if those
need any changes.  Ideally, a spec should be implementable by a person
who only has the spec as his or her guidance as to how to implement a
particular feature.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


RE: Mesa 7.0 in Gutsy

2007-06-28 Thread Johnson, Charles F
What is your position on the new DRM memory management code ?? The
Poulsbo drivers will depend on it. 


Charles Johnson
Ultra-Mobility Group
Platform Software Engineering
Intel Corporation
[EMAIL PROTECTED]
503-712-5181

-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Bryce
Harrington
Sent: Thursday, June 28, 2007 2:15 PM
To: ubuntu-mobile@lists.ubuntu.com
Subject: Mesa 7.0 in Gutsy

Hi all,

Mesa 7.0.0 is now available in Gutsy.  It didn't make the Tribe2 freeze
unfortunately, but will come in once you update.  Please test and let us
know of any issues with it.

Bryce

On Mon, Jun 25, 2007 at 01:34:09AM -0700, Bryce Harrington wrote:
> Mesa 7.0.0 was released Friday.  Test versions of the debs are
available
> here:  http://people.ubuntu.com/~bryce/Testing/mesa_7.0.0/
> 
> Please give them a try; if no serious issues crop up I plan on putting
> them in for upload in the next day or two to be included for Tribe 2.
> 
> Bryce
> 
> -- fwd --
> Date: Fri, 22 Jun 2007 16:57:18 -0600
> From: Brian Paul <[EMAIL PROTECTED]>
> To: MESA-Devel <[EMAIL PROTECTED]>,
> mesa3d-users <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED]
> Subject: [Mesa3d-announce] Mesa 7.0 released
> 
> 
> Mesa 7.0 can be downloaded from SF at
> http://sourceforge.net/project/showfiles.php?group_id=3
> 
> This is a stable release that just fixes bugs since 6.5.3 (see the
> release notes) and makes the OpenGL 2.1 API support official.
> 
> -Brian
> 
> -- 
> Ubuntu-mobile mailing list
> Ubuntu-mobile@lists.ubuntu.com
> Modify settings or unsubscribe at:
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at:
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Mesa 7.0 in Gutsy

2007-06-28 Thread Bryce Harrington
Hi all,

Mesa 7.0.0 is now available in Gutsy.  It didn't make the Tribe2 freeze
unfortunately, but will come in once you update.  Please test and let us
know of any issues with it.

Bryce

On Mon, Jun 25, 2007 at 01:34:09AM -0700, Bryce Harrington wrote:
> Mesa 7.0.0 was released Friday.  Test versions of the debs are available
> here:  http://people.ubuntu.com/~bryce/Testing/mesa_7.0.0/
> 
> Please give them a try; if no serious issues crop up I plan on putting
> them in for upload in the next day or two to be included for Tribe 2.
> 
> Bryce
> 
> -- fwd --
> Date: Fri, 22 Jun 2007 16:57:18 -0600
> From: Brian Paul <[EMAIL PROTECTED]>
> To: MESA-Devel <[EMAIL PROTECTED]>,
> mesa3d-users <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED]
> Subject: [Mesa3d-announce] Mesa 7.0 released
> 
> 
> Mesa 7.0 can be downloaded from SF at
> http://sourceforge.net/project/showfiles.php?group_id=3
> 
> This is a stable release that just fixes bugs since 6.5.3 (see the
> release notes) and makes the OpenGL 2.1 API support official.
> 
> -Brian
> 
> -- 
> Ubuntu-mobile mailing list
> Ubuntu-mobile@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Gutsy Builds & Ubuntu Mobile & Embedded

2007-06-28 Thread Johnson, Charles F
Folks,

I think there was confusion on my part regarding the usefulness (or lack
thereof) of the Gutsy builds
(https://wiki.ubuntu.com/GutsyReleaseSchedule) for MID testing.  My main
question is when do we expect to be able to generate true UME builds ??
Even a target date or Build # would be helpful for planning purposes.  
  

Charles Johnson
Ultra-Mobility Group
Platform Software Engineering
Intel Corporation
[EMAIL PROTECTED]
503-712-5181

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


mobile-graphics spec update

2007-06-28 Thread Johnson, Charles F
Please let me know if more detail is needed.  

Charles Johnson
Ultra-Mobility Group
Platform Software Engineering
Intel Corporation
[EMAIL PROTECTED]
503-712-5181

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Re: Home applet development in Xephyr

2007-06-28 Thread Bryce Harrington
On Thu, Jun 28, 2007 at 10:15:55AM +0200, Johan Bilien wrote:
> Hi,
> 
> On Wed, Jun 27, 2007, Bryce Harrington wrote:
> > I've added the patch in for my next xorg-server upload, which will go
> > into Gutsy after the Tribe2 freeze is over and after I've had a chance
> > to do some testing, probably within the week.
> 
> Thanks, good news. It will also help us here.
> 
> > Have you submitted your patched Feisty version for an SRU yet?
> 
> I don't think it falls in the scope of an SRU. This is not really a bug
> fix, more a change of behavior and semantic. It may be (although it's
> not likely) that some app would rely on the old behavior.

Ahh, sorry, my mistake; I saw the xserver-xephyr packages you pointed to
(http://people.via.ecp.fr/~jobi/xserver-xephyr/) were against Feisty so
made the assumption that this was going into Feisty.

Unfortunately, I've found the latest xorg-server has some build errors
in composite with this patch.  I'll let you know when I've gotten it
fixed up.

Bryce


-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Re: choice of CPU and GNU triplet for lpia

2007-06-28 Thread Matthias Klose
Tollef Fog Heen schrieb:
> Ok, thanks.  Just to verify, those are in the archive now and we can
> then start bootstrapping?

yes. the dpkg change as proposed is now on its way in the archive as well.

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Spec review: USB Client

2007-06-28 Thread Tollef Fog Heen

I am currently in the process of going through the various
specifications submitted for review or proposal.  Since I suspect we
will end up with a couple of emails back and forth, I'll be splitting
those into separate emails.

This email is about the USB client one,
https://wiki.ubuntu.com/MobileAndEmbedded/USBClient

- The release note bit is not filled in, this needs to happen.

- Some boilerplate text is still present in other sections.  Please
  remove anything which isn't applicable to the spec, or fill in the
  section.

The specification lists a couple of different items that need to be
developed:

- Linux utility

- CDC Ethernet

- File-backed storage

- Peripherial controller driver

The spec should at least list how those bits interact and more
importantly, what other features in the OS they hook into and if those
need any changes.  Ideally, a spec should be implementable by a person
who only has the spec as his or her guidance as to how to implement a
particular feature.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Spec review: Graphics

2007-06-28 Thread Tollef Fog Heen

I am currently in the process of going through the various
specifications submitted for review or proposal.  Since I suspect we
will end up with a couple of emails back and forth, I'll be splitting
those into separate emails.

This email is about the Graphics one,
https://wiki.ubuntu.com/MobileAndEmbedded/Graphics

- Use cases are missing

- The design is completely missing; even just something along the
  lines of «we will use a standard DRI based design with the following
  components: A, B, C» will be much better than what is there now.

- What changes does this need in other parts of Ubuntu?  Does it need
  changes to the PCI listing database in discover-data?  Does it need
  changes to the detection in Xorg?  Etc.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Spec review: Image Creation

2007-06-28 Thread Tollef Fog Heen

I am currently in the process of going through the various
specifications submitted for review or proposal.  While approving
specs, I came across one which is exceptional, the Image Creation one
( https://wiki.ubuntu.com/MobileAndEmbedded/ImageCreation ).  It is
well written and comprehensive at the level of being useful as
end-user documentation.

So, I would like to thank Rusty Lynch a lot for his work here.  It's
much appreciated and well done.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Re: choice of CPU and GNU triplet for lpia

2007-06-28 Thread Tollef Fog Heen
* Matthias Klose 

| Tollef Fog Heen schrieb:
|
| > i686 is fine with me, but I suspect the cputable change will then be
| > slightly wrong as it will affect i386 too?
| 
| that's why I'm asking. Currently dpkg-architecture behaves like:
| 
| $ dpkg-architecture -alpia
| dpkg-architecture: warning: Unknown gcc system type i686-linux-gnulp, falling
| back to default (native compilation)
| WAS: i686-linux-gnuDEB_BUILD_ARCH=i386
| DEB_BUILD_ARCH_OS=linux
| DEB_BUILD_ARCH_CPU=i386
| DEB_BUILD_GNU_CPU=i486
| DEB_BUILD_GNU_SYSTEM=linux-gnu
| DEB_BUILD_GNU_TYPE=i486-linux-gnu
| DEB_HOST_ARCH=lpia
| DEB_HOST_ARCH_OS=linux
| DEB_HOST_ARCH_CPU=i686
| DEB_HOST_GNU_CPU=i686
| DEB_HOST_GNU_SYSTEM=linux-gnulp
| DEB_HOST_GNU_TYPE=i686-linux-gnulp

This looks fine to me.

| and for i386:
| 
| $ dpkg-architecture -ai386
| dpkg-architecture: warning: Unknown gcc system type i686-linux-gnulp, falling
| back to default (native compilation)
| WAS: i686-linux-gnuDEB_BUILD_ARCH=i386
| DEB_BUILD_ARCH_OS=linux
| DEB_BUILD_ARCH_CPU=i386
| DEB_BUILD_GNU_CPU=i486
| DEB_BUILD_GNU_SYSTEM=linux-gnu
| DEB_BUILD_GNU_TYPE=i486-linux-gnu
| DEB_HOST_ARCH=i386
| DEB_HOST_ARCH_OS=linux
| DEB_HOST_ARCH_CPU=i386
| DEB_HOST_GNU_CPU=i486
| DEB_HOST_GNU_SYSTEM=linux-gnu
| DEB_HOST_GNU_TYPE=i486-linux-gnu

So does this.

| Is it really necessary to use "gnulp" in the GNU triplet? It's new, and 
*should*
| not break anything because usually everything following linux-* is ignored.

We need to differentiate the GNU triplet at least one place, so yes,
it's necessary to use -gnulp.

| The current GCC packages pick up the triplet provided by dpkg-architecture and
| configure --with-tune=i586 if the debian architecture is lpia.

Ok, thanks.  Just to verify, those are in the archive now and we can
then start bootstrapping?

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Re: choice of CPU and GNU triplet for lpia

2007-06-28 Thread Matthias Klose
Tollef Fog Heen schrieb:
> * Matthias Klose 
> 
> | The last dpkg upload added support for dpkg, setting the cpu used to i386 
> and
> | the GNU triplet to gnulp-linux-i386. Setting the CPU explicitely to i386
> | compared to i586 or i686) will get us other ABIs at least for some packages
> | (GCC); while we can take care of this in the packaging of these few 
> packages, we
> | should use something like i586 or i686 which comes close to the targeted 
> cpu.
> | a proposed patch is attached.
> 
> i686 is fine with me, but I suspect the cputable change will then be
> slightly wrong as it will affect i386 too?

that's why I'm asking. Currently dpkg-architecture behaves like:

$ dpkg-architecture -alpia
dpkg-architecture: warning: Unknown gcc system type i686-linux-gnulp, falling
back to default (native compilation)
WAS: i686-linux-gnuDEB_BUILD_ARCH=i386
DEB_BUILD_ARCH_OS=linux
DEB_BUILD_ARCH_CPU=i386
DEB_BUILD_GNU_CPU=i486
DEB_BUILD_GNU_SYSTEM=linux-gnu
DEB_BUILD_GNU_TYPE=i486-linux-gnu
DEB_HOST_ARCH=lpia
DEB_HOST_ARCH_OS=linux
DEB_HOST_ARCH_CPU=i686
DEB_HOST_GNU_CPU=i686
DEB_HOST_GNU_SYSTEM=linux-gnulp
DEB_HOST_GNU_TYPE=i686-linux-gnulp

and for i386:

$ dpkg-architecture -ai386
dpkg-architecture: warning: Unknown gcc system type i686-linux-gnulp, falling
back to default (native compilation)
WAS: i686-linux-gnuDEB_BUILD_ARCH=i386
DEB_BUILD_ARCH_OS=linux
DEB_BUILD_ARCH_CPU=i386
DEB_BUILD_GNU_CPU=i486
DEB_BUILD_GNU_SYSTEM=linux-gnu
DEB_BUILD_GNU_TYPE=i486-linux-gnu
DEB_HOST_ARCH=i386
DEB_HOST_ARCH_OS=linux
DEB_HOST_ARCH_CPU=i386
DEB_HOST_GNU_CPU=i486
DEB_HOST_GNU_SYSTEM=linux-gnu
DEB_HOST_GNU_TYPE=i486-linux-gnu

Is it really necessary to use "gnulp" in the GNU triplet? It's new, and *should*
not break anything because usually everything following linux-* is ignored.

The current GCC packages pick up the triplet provided by dpkg-architecture and
configure --with-tune=i586 if the debian architecture is lpia.

  Matthias

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Re: IRC meeting Thursday 1600UTC

2007-06-28 Thread Ian
Ola
> = Documentation =
>
>
> Ian Lawrence has taken on the effort of trying to make heads and tails
> of Hildon from a documentation perspective.  There was some discussion as to 
> what documentation
> system to use.  He has set up some preliminary bits on 
> http://proddingthe.net/ which he requests
> feedback on.  He'll also write up a spec about how documentation should be 
> done.
>
> ACTION: Ian Lawrence to create specification about documentation.

I have created a new specification:
https://blueprints.launchpad.net/ubuntu/+spec/ubuntu-mobile-guide

which links through to:
https://help.ubuntu.com/community/UMEGuide

Talking to some guys on ubuntu-doc it seems help.ubuntu.com has a build in 
place to create system
documentation. Keeping the UME Guide here should help prevent fragmentation.

[]'s
Ian

-- 

.''`.
   : :'  :
   `. `'`
 `- Orgulhoso veja -> sudo apt-get install hildon-desktop

http://ianlawrence.info



-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Re: IRC meeting Thursday 1600UTC

2007-06-28 Thread Tollef Fog Heen

Hi,

notes from the IRC meeting last week; sorry about not posting those
before, but I have been on vacation and then busy catching up.

The agenda was as follows:

* Hildon packaging update (Tollef Fog Heen/Adilson Oliveira)
* Flash image creation (Rusty Lynch)
* Kernel and hardware support (Ben Collins)
  - MID Wireless Device Support & Kernel Patches for Thermal
optimizations
* UI, utilities (Bob Spencer)
* Graphics (Charlie Johnson)
* Build infrastructure (Tollef Fog Heen)
* GNOME components (Bob Spencer)
* Documentation
* Ownerless specs:
  - UI guide
  - Browser
  - Media player UI
  - Hardware media decoding
  - USB client
  - Development environment
* Goals for the next week

= Hildon packaging update =

Adilson gave an update on the packaging.  We are now at a point where
a first cut of the packages are in gutsy and you can get a hildon
desktop up and running only using packages in the archive.  This
should make anyone wanting to build applications on top of Hildon able
to do so.  There is a bug in hildon-desktop which makes it crash if
you click the application menu, but we are working on fixing that.
Adilson believed the custom file picker patch for GTK+ had gone absent
and will check up on that.

ACTION: Adilson to check if the GTK+ patch is still present.

= Flash image creation =

The spec is lagging behind the implementation, but the implementation
is quite complete already.  It needs some tweaks to be able to
generate a first generation bootable image, but the Samsung Q1 Ultra
device itself is a challenge due to some hardware without drivers.

= Kernel and hardware support =

There are two bits to this: the necessary kernel patches and the
configuration.  Jacob and Amit are working together on bringing the
patches into the Ubuntu kernel repository, but there are some
minor changes that are still needed.

ACTION: Jacob to clean up patches and work with Amit on getting them
into the kernel repository.

= UI & Utilities =

The specs here are lagging quite badly, but progress on the
implementation during the DAM was good.  Some docs are written, but
not yet put anywhere public, but Bob is going to do that, hopefully by
Monday evening.

ACTION: Bob to put his note and API / UML somewhere for others to
see.

= Graphics =

Postponed, since Charlie had to leave the meeting at the top of the
hour.  Will be followed up by email.

= Build infrastructure =

No progress; Tollef has been on vacation.

= GNOME components =

The spec here is currently empty.  The plan is to fill it with what
parts of GNOME Mobile we want to use.  Depending on whether we want
UME to be a product or a platform, two approaches are possible:

- If we are building a product, just use exactly what our applications
  need.

- If we are building a platform, include GMAE and any extras we need
  for our platform.

Consensus on going for the platform approach was reached.

= Documentation =

Ian Lawrence has taken on the effort of trying to make heads and tails
of Hildon from a documentation perspective.  There was some discussion
as to what documentation system to use.  He has set up some
preliminary bits on http://proddingthe.net/ which he requests feedback
on.  He'll also write up a spec about how documentation should be
done.

ACTION: Ian Lawrence to create specification about documentation.

= Ownerless specifications =

We had a few ownerless specifications who got new owners:

Bob picked up UI guide, Browser, Media player UI, Charlie Johnson got
Hardware media decoding and USB client while Tollef Fog Heen picked up
Development environment

= Goals for next week =

The goal is to have all specifications pushed through to approved by
the next meeting.  Rusty also wants to have the necessary init scripts
in place so it's possible to boot straight into a maemo environment.

Next meeting is Friday at 1600 UTC to accomodate for Rusty travelling
on Thursday.

-- 
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile


Re: Home applet development in Xephyr

2007-06-28 Thread Johan Bilien
Hi,

On Wed, Jun 27, 2007, Bryce Harrington wrote:
> I've added the patch in for my next xorg-server upload, which will go
> into Gutsy after the Tribe2 freeze is over and after I've had a chance
> to do some testing, probably within the week.

Thanks, good news. It will also help us here.

> Have you submitted your patched Feisty version for an SRU yet?

I don't think it falls in the scope of an SRU. This is not really a bug
fix, more a change of behavior and semantic. It may be (although it's
not likely) that some app would rely on the old behavior.

BR,
-- 
Johan Bilien

-- 
Ubuntu-mobile mailing list
Ubuntu-mobile@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-mobile